Coal Plant O&M
To optimize performance,
begin at the pulverizers
A systematic, performance-driven maintenance program for optimizing com-
bustion can achieve great results. The challenge for an O&M staff is de-
ciding which proven strategy and tactics for reducing NOx and improving
plant reliability to adapt and implement. The structured approach pre-
sented here has proven its worth at several plants that have wrestled with
problems similar to yours.
By Richard F. (Dick) Storm, PE, and Stephen K. Storm, Storm Technologies Inc.
O
ptimizing combustion in pulverized
coal (PC)-fired boilers today is more
important today than ever. It is well
known that the average American PC plant
is over 30 years old and that over its life-
Change is in the air
Tighter pollution control has been a
common thread in the major evolution-
ary development of coal-fired genera-
tion over the past 30 years:
n Low-NOx burners, overfire air sys-
tems, and other “furnace solutions”
have enabled major reductions in NOx
emissions, from 0.5 to 1.5 lb/million
Btu to less than 0.15 lb/million Btu.
n Many boilers designed to fire eastern
and midwestern bituminous coals
have been adapted to burn lower-
sulfur Powder River Basin coal at re-
duced production costs.
n Pulverizers designed for coals with a
Hardgrove Grindability Index (HGI) of
50 to 60 are today working with coals
that have an HGI in the low 40s.
n Additions of electrostatic precipita-
tor fields and back-end retrofits of
baghouses, selective catalytic and
noncatalytic reduction systems, and
scrubbers have become common-
place.
n Distributed control systems and ad-
vanced electronic hardware and soft-
ware have modernized and optimized
boiler operations.
n Public and regulatory pressures are
leaning toward mandatory CO2 emis-
sions caps.
|
February 2007 POWER www.powermag.com 37
time NOx and SO2 emissions limits have
been steadily ratcheted down (see box).
Today, operators no longer wonder whether
permissible levels will continue to fall but,
rather, when and by how much.
The newest fork in coal-fired genera-
tion’s path forward is determining how to
capture plant emissions of carbon dioxide
(CO2) when—not if—the gas is regulated
as a pollutant. Some advocate widespread
installations of unproven integrated gasifi-
cation combined cycle (IGCC) technology
ASAP, to prepare it as a long-term solution.
Others say building fleets of super-efficient
supercritical and ultrasupercritical-pressure
and -temperature plants would be a time-
lier, more prudent, and more cost-effective
alternative. But while regulators, Congress,
and the courts wrestle with the question of
what to do about greenhouse gases, one thing
remains clear: CO2 emissions could be low-
ered considerably by raising the efficiency of
the existing U.S. fleet of 1,100+ coal plants.
Today’s average U.S. PC-fired plant oper-
1. Room for improvement. The heat rate of most older coal-fired steam plants could
be lowered by improving their combustion air and fuel systems. Source: Storm Technologies
11,000
10,500
Opportunity for improvement by performance-driven
10,000
9,500
Heat rate (Btu/kWh)
9,000
8,500 8,210
8,000
Station #1 design
7,500
Eddystone
7,000
6,500
6,000
1960
ates at a heat rate of about 10,500 Btu/kWh.
Yet a subcritical (2,400 psi/1,000F/1,000F)
unit is capable of operating at least 10%
more efficiently, at a heat rate of 9,500 Btu/
kWh (Figure 1). There are many proven
ways to improve a boiler’s performance by
continuously optimizing its controllable
variables (see the box on page 38). This
article explores opportunities for raising a
unit’s efficiency by improving the perfor-
mance of its pulverizers.
Storm Technology’s experience has dem-
onstrated that, of the 20 key O&M control-
lable variables with the greatest impact on
unit heat rate (see the box on page 40),
most involve the furnace’s “burner belt.”
Essentially (and most often), in a plant op-
erating at its lowest possible heat rate, the
combustion airflows, pulverizers, fuel line
balancing, burners, and air heaters will all
be optimized.
Despite all the changes in regulations,
equipment, fuels, and combustion controls
over the past few decades, one thing has
10,500
Approximate average for entire U.S. fleet
maintenance of boiler controllable variables
9,500
of large pulverized coal units
8,800
Heat rate capability of typical
8,500 8,540
2,400 psi/1,000F/1,000F
Typical “best” supercritical
subcritical unit
units with low-NOx
burners “actual”
top 10 plants
supercritical
unit designs
Typical new
Average of
1968 1995 2003 2005 2006
Year
 Coal Plant O&M
not changed in evaluating pulverizer per-
formance: You need to get the inputs right!
Table 1 breaks down the potential heat
rate improvements achievable from giving
your pulverizer and related systems a good
tune-up.
13 ways to optimize
combustion of boilers
with low-NOx burners
1. The excess-oxygen level at the fur-
nace exit must be 2% minimum
(and preferably 3%).
2. The fuel lines to the burners should
be balanced by a “clean air” test
accurate to ±2% or better.
3. The fuel lines should be balanced
by a “dirty air” test (using a high-
velocity probe) accurate to ±5% or
better.
4. The flow rates of fuel lines should
be within 10% of each other.
5. At least 75% of the fuel particles
in the lines should be fine enough
to pass through a 200-mesh screen.
Less than 0.1% of the particles
should be able to pass a 50-mesh
screen.
6. Primary airflow should be measured
and controlled with an accuracy of
±3%.
7. Overfire air should likewise be
measured and controlled at ±3%
accuracy.
8. Primary air/fuel ratio should be
accurately controlled when above
minimum.
9. The minimum velocity within fuel
lines should be 3,300 feet per
minute.
Reversal of fortune
Before moving on to some prescriptions for
pulverizer optimization, consider the internal
configuration of a typical, 1970s-vintage 600-
MW PC-fired boiler (Figure 2). The top of
Table 1. Pulverizer improve-
ments are significant and easy to
accomplish. Source: Storm Technologies
Controllable O&M variable
Air in-leakage (reduce)
Primary airflow (optimize)
Pulverizer performance, fuel line
balance (optimize)
Air heater leakage (reduce)
Coal “pyrites” rejects (reduce)
LOI (minimize)
Desuperheating spray water
flows (minimize)
Total
2. Burned to a crisp. In a typical “thirty-something” pulverized coal-fired boiler, 99+% of
combustion should be complete by the time the products of the process reach the convection
surfaces. Source: Storm Technologies
Point at which combustion
should be completed
Residence time
approximately 1-2 secs.
the burner belt is about 55 feet below the nose
arch, or furnace exit. For optimal production
and environmental performance, combustion
must be 99% complete by the time its prod-
ucts are passing over the superheater and
reheater surfaces. The flue gases are moving
swiftly; the residence time in the furnace of
the primary air/coal mixture that entered the
furnace at the top burner level is less than 2
seconds. Usually, residence time from the top
Heat rate
improvement
burners to the nose arch is more like 1 second.
Remembering the shortness of this interval is
potential important when it is essential to minimize
(Btu/kWh) NOx emissions and when firing fuels with
200 high levels of iron and sodium in their ash.
Common in-furnace NOx-reduction so-
50
lutions include using low-NOx burners and
100 overfire air (OFA) systems to intentionally
stage or slow down combustion. Realizing
80
that delayed combustion is fundamental to the
40 design of all low-NOx burners and OFA sys-
80 tems helps in understanding why it is so im-
50 portant to optimize pulverizer performance.
For example, the ash-softening temperature
600 of a bituminous coal whose ash is high in iron
Flame quench zone

10. Mechanical tolerances of burners

and dampers should be ±1/4 inch
or better.
11. Secondary air distribution to burn-
ers should be controlled with an ac-
curacy of between ±5% and ±10%.
12. The fuel feed to pulverizers should
be smooth during load changes and
should be measured and controlled
as accurately as possible, prefera-
bly by gravimetric feeders equipped From forced-draft fan To stack
with load cells.
13. Fuel feed quality and size should be
consistent.

38 |
POWER February 2007
 Coal Plant O&M

20 boiler variables that can be controlled by O&M practices to improve unit

heat rate
1. Flyash loss-on-ignition (LOI), or unburned carbon in ash.
2. Bottom ash carbon content.
3. Boiler and ductwork air in-leakage.
4. Primary airflow. (Measure and control more precisely, to re-
duced tempering airflows.)
5. Pulverizer air in-leakage on suction-fired mills. (Reduce it.)
6. Pulverizer throat size and geometry. (When optimized, reduc-
es coal rejects and complements operation at lower primary
airflows, which reduces tempering airflow and total airflow
bypassing the secondary air heater.)
7. Secondary airflow. (When measured and controlled more
closely, it enables more precise control of furnace stoichiom-
etry—essential to low-NOx operation.)
8. Peak upper furnace exhaust gas temperatures. (When too
high, they foster “popcorn ash” carryover into the selective
catalytic reduction system and air preheater, excessive spray
water flows, and boiler slagging and fouling.)
9. Desuperheating spray water flow to the superheater. (Reduce
the level.)
10. Desuperheating spray water flow to the reheater. (Reduce the
level.)
11. Air heater leakage. (Reduce it; the level for Ljungstrom re-
generative air heaters should and can be less than 9%.)
12. Superheater outlet steam temperature.
13. Reheater outlet steam temperature.
14. Air heater exit gas temperature. (Correct it to a “no leakage”
basis and optimize it.)
15. Burner “inputs” tuning. (For lowest possible excess oxygen at
the boiler outlet and satisfactory NOx and LOI levels.)
16. Boiler exit (economizer exit) gas temperatures ideally be-
tween 650F and 750F, with zero air in-leakage (no dilution).
17. Cycle losses due to valve leak-through. (For spray water
valves, reheater drains to the condenser and superheater,
reheater drains and vents, and—especially—any low-point
drains to the condenser or hotwell.)
18. Sootblowing frequency. (Optimized for maximum cleaning ef-
fect and minimal impact on heat rate.)
19. Steam purity. (Turbine deposits negatively impact unit heat
rate and capacity.)
20. Auxiliary power consumption. (Minimize it by optimizing
fan clearances, duct leakage, fuel and primary air system
performance.)

content may be as much as 300 degrees F lower in a reducing (0%

oxygen) atmosphere than in an oxidizing environment.
Storm Technologies has found that when coal with high sulfur and
high iron content is fired with non-optimized inputs, excessive slag-
ging occurs in the furnace due to the combination of coal chemistry
and secondary combustion. The key point to be made here is that
the effect of chemistry kicks in when the ash becomes “sticky” or
molten, and that happens at a lower temperature with fuel ash high
in iron content.
Low-NOx combustion deliberately consumes some of the furnace
residence time for staging combustion and, as a result, contributes to
more zones in the upper furnace being in an oxygen-deficient state.
Consequently, more slagging occurs and more sootblowing is needed
to remove the slag, which reduces heat transfer. Because increased
sootblowing increases tube erosion and shortens tube life, suboptimal
combustion contributes to reduced plant reliability and availability.

Double down
Suboptimal combustion also takes its toll on unit heat rate:

n Each extra sootblowing cycle imposes an overhead cost on steam

cycle efficiency.
n The carryover of cinders into the air heater increases draft losses
and fan auxiliary power consumption. The increased air heater dif-
ferential will then increase air in-leakage due to the fouling.
n In units retrofitted with low-NOx burners or an OFA system, com-
bustion may be actively occurring higher in the furnace, creating
secondary combustion. This elevated center of combustion will
decrease waterwall heat absorption, elevating the peak furnace
exit gas temperature (FEGT). High FEGTs lower combustion ef-
ficiency and raise unit heat rate.

Given the extremely short furnace residence times and the staging
or slowing of combustion to reduce NOx formation at the source, it
CIRCLE 19 ON READER SERVICE CARD

40 |
POWER February 2007
 Coal Plant O&M
is clear that combustion efficiency must be 3. Stand in the corner. Corner-fired boilers utilize more of the total furnace for combus-
maximized in the burner belt. Unnecessarily tion and tend to have lower NOx emissions. Source: Storm Technologies
high FEGTs can overheat superheater and re-
heater metals and cause higher-than-optimal
desuperheating spray water flows (imposing a
sizable heat rate penalty). Optimizing burner
belt performance requires more precise mea-
surement of key boiler performance variables
and tighter control of the fuel:air ratio. Nose arch

Focus on firing
Carbon
By now, you’re probably asking,“What does
burnout area
low–heat rate, low-NOx boiler performance
have to do with pulverizer operation?” Plen- Separated
ty, as it turns out. Several significant factors overfire air
involved in optimizing combustion with Air-rich zone
Windbox
low-NOx burners are equally applicable to
corner-fired and wall-fired boilers.
Let’s first consider corner-fired boilers,
which are considered inherently forgiving of Fuel-rich zone
less-than-optimal combustion tuning. This
tolerance seems to derive from the fact that
in corner-fired boilers (Figure 3), the entire
furnace volume can be considered a single
burner into which fuel and air are injected
from the corners, creating a burning mass in
the center. The burning mass serves to reduce
peak temperatures. Meanwhile, tangential
injections from the corners impart “swirl”
to the fuel and combustion air at all burn-
ers; that creates a more homogeneous mix
of the products of combustion for fuel-rich
and fuel-lean burners. The tangential admis-
4. Over and above. NOx emissions can be reduced by applying high-momentum overfire
sion of the fuel and air also slightly increases
air through opposed nozzles in the upper furnace. Source: Storm Technologies
the residence time of a coal particle, from its
introduction into the furnace until its conver-
sion to a burning carbon char particle that is
quenched to below 1,400F in the convection
pass.
The “burning mass” principle also reduc-
es the intensity of combustion in the burner
belt, lowering “natural” NOx levels in the
process. Before low-NOx burner technology
was perfected, corner-fired boilers inherent-
ly produced less NOx than wall-fired boilers.
Modern, low-NOx designs stage combustion
not only vertically, but horizontally as well.

Learn from experience Overfire air zone

Another proven way to reduce NOx emis- to complete
sions without imposing a heat rate penalty is combustion
Stoichiometry = 1.15
to apply high-momentum OFA through op-
posed nozzles in the upper furnace. Figure
4 shows the configuration of such a system,
as installed both at AES Corp.’s 126-MW
Westover Plant in Johnson City, N.Y. (see
Burner zone
POWER, October 2006, p. 26) and Savannah Stoichiometry ≤1.00
Electric’s McIntosh Plant.
In Storm Technologies’ experience, wall-
fired boilers require fine-tuning of the fuel
and air inputs to the burner belt to opti-
mize plant performance, response, and heat
rate. A good example is McIntosh Unit 1, a

|
February 2007 POWER 41
 Coal Plant O&M
5. You’re fired. McIntosh Plant Unit 1 before (L) and after (R) retrofit of a fan-boosted overfire air system. Source: Storm Technologies
1975-vintage, 175-MW unit with 16 front
wall-mounted burners (Figure 5, left). The
furnace division panel separated the eight
burners on the right side of the boiler from
the eight burners on the left. The 16 burn-
ers are arranged in four levels of four across.
A fan-boosted OFA system was retrofitted
to the unit (Figure 5, right) with excellent
results (Table 2). For McIntosh Unit 1, the
overall project included optimization of its
pulverizers, burners, combustion airflow
paths (primary, secondary, and OFA), and
improvements to the management and con-
trol functions of the systems.
This wall-fired unit—which has an unfor-
giving furnace arrangement (furnace divi-
sion panels and wall-mounted burners) and
a relatively short residence time (1 second
from the top burners to the nose arch)—
validates the potential of using the tenets of
combustion optimization found in Table 1. It
has test-burned a wide range of coals from
South America, Central Appalachia, and the
Powder River Basin with good results.
Those results include improved reliability,
which must be quantified and factored into
the savings equation. Optimized combustion
has reduced slagging and fouling. Improved
fuel fineness, fuel distribution, and combus-
tion air distribution also have contributed to
greater unit availability.
One less well-documented advantage of
greater fuel fineness (75% of coal particles
pass a 200-mesh screen, and none pass one
42
Table 2. Results of retrofitting a fan-boosted overfire air system to
Unit 1 of Savannah Electric’s McIntosh Plant. Source: Storm Technologies
Variable Pre-retrofit Post-retrofit
Coal fineness (percentage of <HGI coal 50%–60% 75%–80%
particles passing 200-mesh screen )
Clean airflow balance (distribution 5%–17% <2%
imbalances)
Fuel flow balance (distribution ±20% ±10%
imbalances)
Air-fuel ratio (lb air/lb fuel) >2.0 ~1.8
Flyash loss-on-ignition with Eastern 16%–22% <10%
bituminous coals
Flyash loss-on-ignition with low-rank Not available <15%
Venezuelan coals
Furnace exit gas temperature (average ~2,250F/2,400F <2,200F/2,300F
FEGT/maximum FEGT)
NOx (lb/mmBtu) 0.78–1.0 ~0.28/0.36 (with 3 or 4 mills operating)
of 50 mesh) is reduced waterwall wast- 30+-year-old plants install SO2 scrubbers
age. This type of corrosion becomes more and then are converted to be capable of fir-
severe on boilers operating at supercritical ing higher-sulfur coals. Why? The reason is
pressure and firing fuels with high sulfur because most low-sulfur compliance coals
and iron content. Even boilers running at also have low iron content in their ash.
1,600 to 1,800 psi can have their useful life These coals are “forgiving” from the per-
shortened considerably if their waterwalls spective of contributing to aggressive fire-
are exposed to highly aggressive fireside side corrosion of water walls and slagging.
corrodents of sulfur and iron in a reducing Once a plant owner has spent the money
environment. on a scrubber, he will be tempted to buy a
higher-sulfur fuel (with higher iron content
Future shock in its ash) for economic reasons—primarily
In fact, the industry can expect fireside escalating Western coal rail costs and im-
wastage to become more prevalent as more proved Eastern coal cost-competitiveness.
|
POWER February 2007
 Coal Plant O&M
6. Walk the line. Economic plant operation requires operation in the narrow zone of
optimum combustion efficiency. Source: Storm Technologies

DELIVERS! Loss Save Loss

March 2007
Unburned Zone of Operating here

Heat lost in flue gases (%)

� Special Report: Design and
O&M of combined-cycle plants
� Using gray water for condens-
er cooling
� Preventing water hammer in
steam systems
� Powering up energy equipment
exports
� Harvey Research Issue: Ad-Q
advertising readership study
Closing Date:
February 22, 2007
April 2007
� ELECTRIC POWER show issue
� Nuclear plant plans move
forward
� Next-generation mercury moni-
tors reviewed
� Solving lubrication problems
Closing Date:
March 23, 2007
fuel loss
combustion
efficiency
Operating here
results in
wasted fuel
Total air (%)
This “reverse fuel switch” trend is likely to
foster more slagging, fouling, and aggres-
sive furnace tube corrosion.
Fortunately, there are options available
to prevent increases in slagging and fouling
that result from a reverse fuel switch. One
for minimizing fireside wastage is to opti-
mize pulverizer performance to fuel fine-
ness that’s acceptable in all fuel lines at all
times. Storm Technologies’ standard mini-
mum recommended fineness is greater than
75% of particles passing through a 200-
mesh screen and none through a 50-mesh
screen. Lowering superheater and reheater
metal temperatures, to reduce slagging and
fouling in the convection passes, is another
step than can be taken to improve unit reli-
ability and burner belt combustion.
Finally, operating a unit at its maximum
efficiency and capacity should be an over-
riding economic objective. Optimizing pul-
verizer performance and burner belt inputs
can help reach that goal. So can diagnostic
performance testing of fuel lines, combus-
tion airflows, and key upper furnace com-
bustion parameters. When fuel line fineness
declines, and reducing environments are
found in the furnace, corrective action
should be taken immediately.
Cleaner coal-burner
optimum results in excess heat
lost up the stack
Excess air loss
ground. Poor fineness traps fuel-bound
nitrogen within the carbon char particles,
beyond the reach of even the best low-NOx
burners.
Two. Fuel balance usually improves
with better fuel fineness. A powdery mix-
ture of fine coal particles passing through
a pulverizer, classifier, and coal riffles (if
the unit is so equipped) will flow more uni-
formly when it is more finely ground. Such
fuels actually flow more like a gas when
entrained at the proper ratio in the primary
air stream. Fine coal particles mixed in the
transport air become more uniformly dis-
tributed than coarsely ground coal particles
at a similar air/fuel ratio.
Three. A more homogeneous mixture of
coal and air entering the burners will natu-
rally reduce required excess air levels. By
reducing the total airflow and reducing the
excess air, thermal NOx production is re-
duced. The better the mixing of the combus-
tion products in the available residence time,
the less “extra air” that has to be added to
create oxidizing zones in portions of the fur-
nace that are fuel-rich (Figure 6). When these
“peaks and valleys” of free oxygen and high
temperatures are made more uniform, then it
becomes possible to reduce the total excess
air that has to be added to make up for impre-
cise fuel and air inputs into the burner belt. n

ACT NOW! To sum up, there are three significant ways

in which optimizing pulverizer performance
—Dick Storm is president of Storm
Technologies (www. stormeng.com);
Reserve Your Ad can contribute to a reduction in a coal-fired
boiler’s NOx emissions.
he can be reached at (704) 983-2040 or
skstorm1@aol.com. Stephen K. Storm is
Contact your sales One. Release of fuel-bound nitrogen a vice president of the company and its
representative in the burner’s devolatilization zone is en-
hanced by making coal particles smaller,
manager of technical field services;
he can be reached at 704-983-2040
(listed on p. 2) in effect increasing the overall fuel surface or skstorm1@aol.com.
area. Low-NOx burners are most effective
when they’re fed coal that has been finely

44 |
POWER February 2007